Cyboflow rev 0014 · features Home Features Download
Capabilities

Everything in the cockpit.

Six systems for running a fleet of coding agents with humans at the center — dispatch the work, route your attention to what matters, and tune the whole thing as it learns.

Feature 01 · Custom workflows

Design and run custom multi-step workflows.

Mirror the same systems that have made human development reliable — planning, code review, testing, verification — for an agentic environment.

  • Compose plan, execute, verify, and review into one flow.
  • Customize your agents.
  • Save flows as reusable templates.
soloflow.yaml · running
PLAN
✓ done
EXEC
• running
VERIFY
pending
REVIEW
pending
human review · 3 pending
permexecutor wants to run BashApprove
decwhich auth provider to wire?Decide
actreview diff · +156 −114Open
Feature 02 · Human review queue

Everything that needs your input, in one place.

A central view for all your running agents, grouped by whether the input needed is a permission, a decision, or an action.

  • One queue across every running agent.
  • Approve, deny, or steer without leaving the page.
  • Launch the next agent the moment one unblocks.
Feature 03 · Observability

Measure your system, tune what matters.

Built-in observability across every run — track token use, error and bug rates, and throughput. See where the waste is, then tune a leaner, more efficient system.

  • Measure token usage, run failures, bugs and issues.
  • Track performance by workflow, agent, and project.
  • Compare workflow versions over time so you can tune your system.
observability · live
tokens / run
128k↓18%
run error rate
0.4%↓2.1pt
post merge bugs
12/wk↓60%
task throughput
34/d↑41%
learning · findings
12
pending
107
resolved
0
dismissed
Executor re-ran the failing suite 4×
Quick fix
ApproveDismiss
verify step skips lint on docs-only diffs
Documentation update
ApproveDismiss
Plan under-scoped the auth refactor
Task candidate
ApproveDismiss
Feature 04 · Learning

Help your system get better over time.

Build in learning loops, so your system improves itself — every completed run feeds the next one.

  • Agents issue findings — problems, missing tests, documentation gaps, opportunities for improvement
  • A dedicated home to triage those findings so you can decide which ones to act on
  • A built in compounding loop to implement those improvements and tighten your whole system
Feature 05 · SDK or PTY

Your terminal, or our scaffolding.

Built with native support for Claude Code — run Cyboflow in the terminal you already know, or reach for the Cyboflow scaffolding when the job gets hard.

  • Native Claude Code support, out of the box.
  • Drive it as a PTY in the terminal you already use.
  • Or call the SDK and wrap the hardest jobs in Cyboflow scaffolding.
zsh — claude code
PTYSDK
› ~/app claude
  cyboflow attached · soloflow harness
▸ dispatch   executor · opus-4.5
✓ plan       12 steps · 3 phases
▸ execute    editing app/wizard.py
✓ verify     tests green
▸ review     waiting on human

Stop being a babysitter, start becoming a manager.

The control plane for agentic development. We’re onboarding teams who already run coding agents and are tired of hand-holding, constant pings, and confident garbage.